Summary of From Explicit Rules to Implicit Reasoning in An Interpretable Violence Monitoring System, by Wen-dong Jiang et al.
From Explicit Rules to Implicit Reasoning in an Interpretable Violence Monitoring System
by Wen-Dong Jiang, Chih-Yung Chang, Ssu-Chi Kuai, Diptendu Sinha Roy
First submitted to arxiv on: 29 Oct 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed RuleVM framework combines implicit and explicit knowledge to design expert-driven and interpretable violence surveillance systems. Building upon pre-trained models, RuleVM uses a dual-branch structure for images and text. The implicit branch focuses on coarse-grained binary classification using visual features, while the explicit branch leverages language-image alignment for fine-grained classification. This framework achieves interpretable violence surveillance through extensive experiments on two benchmarks, significantly outperforming existing state-of-the-art methods. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary A new way to detect and track violence in videos is being developed. It’s called RuleVM and it uses a special computer program that looks at both pictures and words. One part of the program, called the implicit branch, looks at pictures to say if there’s violence or not. The other part, called the explicit branch, looks at what people are saying in the video to get more details. This new way is better than old ways because it can explain why it thinks there’s violence and it’s very good at finding it. |
Keywords
» Artificial intelligence » Alignment » Classification
